【视频开发】【计算机视觉】doppia编译之四：安装其他库、编译和运行doppia
（与本节内容无关///////////////////////////保存图片参数为—-gui.save_all_screenshots true////////////////////）
在我们安装好CUDA、boost、OpenCV之后，接下来的一些库（libSDL、protobuf等）的安装，我们都可以用系统内部的程序进行安装。比如
安装libSDL,我们终端输入
apt-cache search libsdl
- 1
- 1
系统会给出一系列程序，我们选择其中的libsdl1.2-dev进行安装。
sudo apt-get install libsdl1.2-dev
- 1
- 1
(这里注意不要安装libsdl2-dev,因为安装之后生成的文件夹是SDL2,之后doppia调用时会出现找不到“SDL files”的错误)
安装protobuf库，则是我经过多番测试，得到的能够通过v1,v2测试的安装方法。（之前尝试安装protobuf2.5.0和protobuf2.4.1,doppia都找不到路径，而想把它们删除又删除不了，很麻烦），之后我测试了几个自带的protobuf库，发现安装以下四个库能够通过v1,v2的测试，安装命令为：
sudo apt-get install libprotobuf-dev libprotoc-dev python-protobuf protobuf-compiler
- 1
- 1
切换到doppia目录下，运行
sudo sh ./generate_protocol_buffer_files.sh
- 1
- 1
protobuf通过doppia-v1检测的返回信息为
Generating objects detection files...
(Ground plane and video input files not yet handled by this script)
End of game. Have a nice day!
- 1
- 2
- 3
- 1
- 2
- 3
protobuf通过doppia-v2检测的返回信息为
+ cd src/objects_detection/
+ protoc --cpp_out=./ detector_model.proto detections.proto
+ protoc --python_out=../../tools/objects_detection/ detector_model.proto detections.proto
+ cd ../..
+ cd src/stereo_matching/ground_plane/
+ protoc --cpp_out=./ plane3d.proto
+ protoc --python_out=../../../tools/stixels_evaluation plane3d.proto
+ cd ../../..
+ cd src/stereo_matching/stixels/
+ protoc --cpp_out=./ -I. -I../ground_plane --include_imports stixels.proto ground_top_and_bottom.proto
--include_imports only makes sense when combined with --descriptor_set_out.
+ protoc --python_out=../../../tools/stixels_evaluation -I. -I../ground_plane --include_imports stixels.proto ground_top_and_bottom.proto
--include_imports only makes sense when combined with --descriptor_set_out.
+ cd ../../..
+ cd src/video_input/calibration
+ protoc --cpp_out=./ calibration.proto
+ cd ../../..
+ cd src/helpers/data
+ protoc --cpp_out=./ DataSequenceHeader.proto
+ protoc --python_out=../../../tools/data_sequence DataSequenceHeader.proto
+ cd ../../..
+ cd src/helpers
+ cd ../..
+ cd src/tests/data_sequence/
+ protoc --cpp_out=./ TestData.proto
+ cd ../../..
+ echo End of game. Have a nice day!
End of game. Have a nice day!
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
到这里，我们该安装的库大部分已经安装成功，接下来就可以开始编译doppia啦！（至于可能还缺少的库，可以根据doppia的错误提示进行安装）
编译运行doppia/src/applications/objects_detection
由于我只需要用到doppia的objects_detection的功能，而之前我在编译doppia-v2时，ground_estimation和stixel_world都能编译运行。所以这次在编译doppia-v1时，我就直接切入“主题”，编译运行objects_detection。下面也主要是列出我在编译objects_detection是遇到的问题以及相应的解决方案。
错误一，创建(build)错误
error:
/home/mx/doppia/src/applications/objects_detection/../../../src/helpers/data/DataSequence.hpp:293:56: error: invalid use of incomplete type ‘class google::protobuf::io::CodedInputStream’
const bool read_size_success = input_coded_stream_p->ReadLittleEndian64(&size);
- 1
- 2
- 1
- 2
solution:
在
doppia/src/helpers/data/DataSequence.hpp
- 1
- 1
头文件中，在
#include "DataSequenceHeader.pb.h"
#include <google/protobuf/io/zero_copy_stream_impl.h>
- 1
- 2
- 1
- 2
两行之间添加一行新的引用，如下，
#include "DataSequenceHeader.pb.h"
#include <google/protobuf/io/coded_stream.h>
#include <google/protobuf/io/zero_copy_stream_impl.h>
- 1
- 2
- 3
- 1
- 2
- 3
错误二，创建(build)错误
error:
/home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:23:9: error: ‘fast_stage_t’ in ‘class doppia::SoftCascadeOverIntegralChannelsModel’ does not name a type
typedef SoftCascadeOverIntegralChannelsModel::fast_stage_t cascade_stage_t;
^
/home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:36: error: ‘cascade_stage_t’ was not declared in this scope
typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t;
^
/home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:51: error: template argument 1 is invalid
typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t;
^
/home/mx/doppia/src/applications/objects_detection/../../../src/objects_detection/gpu/integral_channels_detector.cu.hpp:33:86: error: invalid type in declaration before ‘;’ token
typedef Cuda::DeviceMemoryLinear2D<cascade_stage_t> gpu_detection_cascade_per_scale_t;
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
这里出现错误原因是因为common_settings.cmake中没有添加cuda链接库路径。
solution:
编辑common_settings.cmake,在其中添加一个条件项
elseif(${HOSTNAME} STREQUAL "mx-pc")
message(STATUS "Using mx-pc optimisation options")
option(USE_GPU "Should the GPU be used ?" TRUE)
set(CUDA_BUILD_CUBIN OFF)
set(local_CUDA_LIB_DIR "/usr/local/cuda/lib64")
set(cuda_LIBS "")
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 1
- 2
- 3
- 4
- 5
- 6
- 7
这里mx-pc是我电脑的主机名，你需要将它改成自己电脑的主机名。
错误三，创建(build)错误
error:
/home/mx/doppia/src/objects_detection/SoftCascadeOverIntegralChannelsFastFractionalStage.cpp:24:9: error: ‘swap’ is not a member of ‘std’
std::swap(weak_classifier.level2_true_node, weak_classifier.level2_false_node);
- 1
- 2
- 1
- 2
solution:
在
doppia/src/objects_detection/SoftCascadeOverIntegralChannelsFastFractionalStage.cpp
- 1
- 1
文件开头添加一行引用
#include<iostream>
- 1
- 1
解决了以上三个错误后，doppia就可以创建(build)成功啦！但要想运行成功，还得改正以下2个错误。
错误四，链接(link)错误
error:
Linking CXX executable objects_detection
/usr/bin/ld: cannot find -lboost_program_options-mt
/usr/bin/ld: cannot find -lboost_filesystem-mt
/usr/bin/ld: cannot find -lboost_system-mt
/usr/bin/ld: cannot find -lboost_thread-mt
collect2: error: ld returned 1 exit status
make[2]: *** [objects_detection] 错误 1
make[1]: *** [CMakeFiles/objects_detection.dir/all] 错误 2
make: *** [all] 错误 2
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
这里出现错误的原因，是boost库链接出错，这时候我们需要修改CMakeList.txt文件，这里我就直接把CMakeList.txt贴出来，修改的地方做过注释。
solution:
# This is a CMake build file, for more information consult:
# http://en.wikipedia.org/wiki/CMake
# and
# http://www.cmake.org/Wiki/CMake
# http://www.cmake.org/cmake/help/syntax.html
# http://www.cmake.org/Wiki/CMake_Useful_Variables
# http://www.cmake.org/cmake/help/cmake-2-8-docs.html
# to compile the local code you can use: cmake ./ && make -j2
# ----------------------------------------------------------------------
# Base CMake setup
cmake_minimum_required (VERSION 2.6)
set(doppia_root "../../..")
set(CMAKE_MODULE_PATH $ENV{CMAKE_MODULE_PATH})
set(CMAKE_MODULE_PATH "./" ${doppia_root} ${CMAKE_MODULE_PATH})
set(CMAKE_MODULE_PATH "/home/rodrigob/work/code/doppia_references/cuda/FindCUDA/CMake/cuda" ${CMAKE_MODULE_PATH})
set(CMAKE_MODULE_PATH "/users/visics/rbenenso/code/references/cuda/FindCUDA/CMake/cuda" ${CMAKE_MODULE_PATH})
# ----------------------------------------------------------------------
# Setup the project
include(FindPkgConfig)
project (ObjectsDetection)
# ----------------------------------------------------------------------
# Site specific configurations
include(${doppia_root}/common_settings.cmake)
# ----------------------------------------------------------------------
# Setup required libraries
pkg_check_modules(libpng REQUIRED libpng)
#pkg_check_modules(OpenEXR REQUIRED OpenEXR)
pkg_check_modules(opencv REQUIRED opencv>=2.3)
#set(vw_LIBRARIES "-lvwCore -lvwImage -lvwStereo -lvwFileIO -lvwMath -lvwInterestPoint")
set(opencv_LIBRARIES
opencv_core opencv_imgproc opencv_highgui opencv_ml
opencv_video opencv_features2d
opencv_calib3d
#opencv_objdetect opencv_contrib
opencv_legacy opencv_flann
) # quick hack for opencv2.4 support
# 修改1:find where is boost
#(+)find_package(Boost REQUIRED
#(+) COMPONENTS program_options filesystem system thread
#(+) )
find_package(Boost REQUIRED
COMPONENTS program_options filesystem system thread
)
# ----------------------------------------------------------------------
# Setup CUDA
if(USE_GPU)
find_package(CUDA 4.0 REQUIRED)
include_directories(${CUDA_INCLUDE_DIRS} ${CUDA_CUT_INCLUDE_DIR})
endif(USE_GPU)
# ----------------------------------------------------------------------
# Setup link and include directories
set(local_LIBRARY_DIRS
"/usr/local/lib"
"/users/visics/rbenenso/no_backup/usr/local/lib"
"/usr/lib64"
"/usr/lib64/atlas"
"/usr/lib/sse2/atlas"
"/usr/lib/llvm-2.8/lib"
${local_CUDA_LIB_DIR}
)
set(local_INCLUDE_DIRS
"/users/visics/rbenenso/no_backup/usr/local/include"
"/usr/include/eigen2/"
"/usr/local/include/eigen2"
"/usr/local/cuda/include"
${CUDA_INCLUDE_DIRS}
)
link_directories(
${libpng_LIBRARY_DIRS}
${OpenEXR_LIBRARY_DIRS}
${opencv_LIBRARY_DIRS}
${local_LIBRARY_DIRS}
)
include_directories(
"${doppia_root}/libs"
"${doppia_root}/src"
${libpng_INCLUDE_DIRS}
${OpenEXR_INCLUDE_DIRS}
${opencv_INCLUDE_DIRS}
${local_INCLUDE_DIRS}
"${doppia_root}/libs/cudatemplates/include"
)
if(USE_GPU)
cuda_include_directories("${doppia_root}/libs/")
endif(USE_GPU)
# ----------------------------------------------------------------------
# Collect source files
set(doppia_src "${doppia_root}/src")
set(doppia_stereo "${doppia_root}/src/stereo_matching")
file(GLOB SrcCpp
"./ObjectsDetection*.cpp"
"./draw*.cpp"
"${doppia_src}/*.cpp"
#"${doppia_src}/objects_detection/*.c*"
"${doppia_src}/objects_detection/Abstract*.c*"
"${doppia_src}/objects_detection/*Converter.c*"
"${doppia_src}/objects_detection/Base*.c*"
"${doppia_src}/objects_detection/*Factory.c*"
"${doppia_src}/objects_detection/Greedy*.c*"
"${doppia_src}/objects_detection/Detection*.c*"
"${doppia_src}/objects_detection/*Model.c*"
"${doppia_src}/objects_detection/*Stage.c*"
"${doppia_src}/objects_detection/*Integral*.c*"
"${doppia_src}/objects_detection/MultiscalesIntegral*.c*"
"${doppia_src}/objects_detection/integral_channels/Integral*.cpp"
"${doppia_src}/objects_detection/FastestPedestrian*.c*"
"${doppia_src}/objects_detection/DetectorSearchRange.c*"
"${doppia_src}/objects_detection/*.pb.c*"
"${doppia_src}/objects_detection/non_maximal_suppression/*.c*"
"${doppia_src}/objects_tracking/*.cpp"
"${doppia_src}/applications/*.cpp"
"${doppia_src}/applications/stixel_world/*Gui.cpp"
"${doppia_src}/applications/stixel_world/draw*.cpp"
#"${doppia_stereo}/*.cpp"
"${doppia_stereo}/cost_volume/*CostVolume.cpp"
"${doppia_stereo}/cost_volume/*CostVolumeEstimator*.cpp"
"${doppia_stereo}/cost_volume/DisparityCostVolumeFromDepthMap.cpp"
"${doppia_stereo}/cost_functions.cpp"
"${doppia_stereo}/CensusCostFunction.cpp"
"${doppia_stereo}/CensusTransform.cpp"
"${doppia_stereo}/GradientTransform.cpp"
"${doppia_stereo}/AbstractStereoMatcher.cpp"
"${doppia_stereo}/AbstractStereoBlockMatcher.cpp"
"${doppia_stereo}/SimpleBlockMatcher.cpp"
"${doppia_stereo}/MutualInformationCostFunction.cpp"
"${doppia_stereo}/ConstantSpaceBeliefPropagation.cpp"
"${doppia_stereo}/qingxiong_yang/*.cpp"
"${doppia_stereo}/SimpleTreesOptimizationStereo.cpp"
"${doppia_stereo}/OpenCvStereo.cpp"
"${doppia_stereo}/ground_plane/*.cpp"
"${doppia_stereo}/stixels/*.cpp"
#"${doppia_stereo}/stixels/*.cc"
"${doppia_src}/video_input/*.cpp"
"${doppia_src}/video_input/calibration/*.c*"
"${doppia_src}/video_input/preprocessing/*.cpp"
#"${doppia_src}/features_tracking/*.cpp"
"${doppia_src}/image_processing/*.cpp"
"${doppia_src}/drawing/gil/*.cpp"
)
file(GLOB HelpersCpp
#"${doppia_src}/helpers/*.cpp"
"${doppia_src}/helpers/data/*.c*"
"${doppia_src}/helpers/any_to_string.cpp"
"${doppia_src}/helpers/get_section_options.cpp"
"${doppia_src}/helpers/Log.cpp"
"${doppia_src}/helpers/loggers.cpp"
"${doppia_src}/helpers/AlignedImage.cpp"
"${doppia_src}/helpers/replace_environment_variables.cpp"
"${doppia_src}/helpers/objects_detection/*.cpp"
)
file(GLOB SrcGpuCpp
"${doppia_src}/objects_detection/Gpu*.cpp"
"${doppia_src}/objects_detection/integral_channels/Gpu*.cpp"
"${doppia_src}/helpers/gpu/*.cpp"
#"${doppia_stereo}/SimpleTreesGpuStereo.cpp"
)
file(GLOB SrcCuda
"${doppia_src}/objects_detection/integral_channels/gpu/*.cu"
"${doppia_src}/objects_detection/integral_channels/gpu/*.cpp"
"${doppia_src}/objects_detection/gpu/*.cu"
"${doppia_src}/objects_detection/gpu/*.cpp"
#"${doppia_src}/helpers/gpu/*.cu"
# "${doppia_stereo}/*.cu.c*"
# "${doppia_stereo}/*.cu"
# "${doppia_stereo}/gpu/*.cu.c*"
# "${doppia_stereo}/gpu/*.cu"
)
list(REMOVE_ITEM SrcCpp ${SrcCuda}) # just in case
if(USE_GPU)
# add GPU related source code to the executable list
list(APPEND SrcCpp ${SrcGpuCpp})
# add GPU related libraries
list(APPEND opencv_LIBRARIES opencv_gpu)
# ----------------------------------------------------------------------
# Compile CUDA stuff
cuda_include_directories(${local_CUDA_CUT_INCLUDE_DIRS})
cuda_include_directories(${CUDA_INCLUDE_DIRS} ${CUDA_CUT_INCLUDE_DIR} ${local_CUDA_CUT_INCLUDE_DIR})
link_directories(${local_CUDA_CUT_LIBRARY_DIRS})
cuda_add_library(cuda_stuff_library ${SrcCuda})
target_link_libraries(cuda_stuff_library
${CUDA_LIBRARIES}
${cutil_LIB}
)
#set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} --generate-line-info) # used during profiling
endif(USE_GPU)
# ----------------------------------------------------------------------
# Create the executable
#set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++0x") # required for unrestricted unions
#set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -p") # add gprof information
add_library(cpp_stuff_library ${SrcCpp} ${HelpersCpp})
add_executable(objects_detection "./objects_detection.cpp")
target_link_libraries(objects_detection
cpp_stuff_library
${cg_LIBRARIES}
# linking with CgGL _after_ boost_program_options generates a segmentation fault ! boost_program_options 1.39 has a bug
#修改2:link to boost
#(-)boost_program_options-mt boost_filesystem-mt boost_system-mt boost_thread-mt
#(+)${Boost_LIBRARIES}
${Boost_LIBRARIES}
protobuf pthread
SDL X11 Xext #Xrandr
gomp
${libpng_LIBRARIES} jpeg
# ${OpenEXR_LIBRARIES}
${opencv_LIBRARIES}
#${vw_LIBRARIES}
#csparse sparse spblas mv
#lapack blas atlas
${google_perftools_LIBS} # enables profiling, see http://code.google.com/p/google-perftools
#`OcelotConfig -l`
#ocelot
#boost_system-mt boost_filesystem-mt boost_thread-mt
#GLEW
#LLVMAsmParser LLVMX86Disassembler LLVMX86AsmParser LLVMX86CodeGen LLVMSelectionDAG
#LLVMAsmPrinter LLVMMCParser LLVMX86AsmPrinter LLVMX86Info LLVMJIT
#LLVMExecutionEngine LLVMCodeGen LLVMScalarOpts LLVMInstCombine LLVMTransformUtils LLVMipa
#LLVMAnalysis LLVMTarget LLVMMC LLVMCore LLVMSupport LLVMSystem
)
if(USE_GPU)
target_link_libraries(objects_detection cuda_stuff_library ${local_CUDA_LIB})
endif(USE_GPU)
# ----------------------------------------------------------------------
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
到这里就OK啦！
最后运行成功你会看到一个简短地video，以及下面这样的信息
2015-05-09 16:33:23 {7feb06974880} [ BaseIntegralChannelsDetector ] : Warning: At scale index 47 the detection window size is larger than the biggest ground plane corridor. Setting the detection search to a single line.
scale_index == 47, original_height == 18, updated_height == 1
Expected speed gain == 5.28x (num pixels original/updated)
GpuVeryFastIntegralChannelsDetector::compute_v2 max search range (min_x, min_y; max_x, max_y) == (0, 0; 153, 58)
2015-05-09 16:33:23 {7feb06974880} [ GpuIntegralChannelsDetector ] : scaled_x == 640, scaled_y == 480
Requested frame number 11 but frames should be in range (0, 10)
Processed a total of 10 input frames
Average objects detection speed per iteration 29.36 [Hz] (in the last 10 iterations)
End of game, have a nice day.
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
好啦，doppia编译到此也就结束啦！
希望这几篇文章能帮助正在读博客的你。
doppia及作者相关介绍链接：
http://blog.csdn.net/xizero00/article/details/43227019
https://bitbucket.org/rodrigob/doppia
【视频开发】【计算机视觉】doppia编译之四：安装其他库、编译和运行doppia的更多相关文章
- windows下编译和安装boost库
boost是一个功能强大.构造精巧.跨平台.开源并且完全免费的C++程序库. 获取方式 boost提供源码形式的安装包,可以从boost官方网站下载,目前最新版本是1.59.0. 本机上正好有boos ...
- [转]Linux编译和安装boost库
1. 下载boost安装包并解压缩 到http://www.boost.org/下载boost的安装包,以boost_1_58_0.tar.gz为例 下载完成后进行解压缩: tar zxvf boos ...
- linux 下的两种软件安装方式 —— 源码（编译、安装），编译好的二进制（直接安装）
我们以 GPG(加密工具)为例来说明两种安装方式的区别: 源码(Source code releases,名称中则会含有src等说明信息,tarball:source),先编译再安装 GPU 的源码地 ...
- CentOS8安装Geant4笔记（三）：Geant4介绍、编译、安装支持Qt5界面并运行exampleB1例程显示Qt界面
前言 上一篇,安装了Qt5环境. 本篇在服务器CentOs8.2上安装geant4软件,geant4使用Qt5来显示. GEANT4 介绍 Geant4 是一个用于模拟粒子穿过物质的工具 ...
- 【视频开发】【计算机视觉】doppia编译之三：编译安装opencv库
这里我介绍2种方法 (1)利用别人写好的脚本编译,相对来说省力一点 上Github下载别人写好的脚本文件,网址 https://github.com/jayrambhia/Install-OpenC ...
- C/C++ 跨平台交叉编译、静态库/动态库编译、MinGW、Cygwin、CodeBlocks使用原理及链接参数选项
目录 . 引言 . 交叉编译 . Cygwin简介 . 静态库编译及使用 . 动态库编译及使用 . MinGW简介 . CodeBlocks简介 0. 引言 UNIX是一个注册商标,是要满足一大堆条件 ...
- 如何安装Eigen库和Sophus库
* { font-family: "Tibetan Machine Uni", "sans-serif", STFangSong; outline: none ...
- SQLite3动态库、静态库编译
资源准备 1.下载SQLite3源码,下载地址为https://www.sqlite.org/download.html.下载sqlite-amalgamation-3200000.zip和sqlit ...
- iOS - 编译WebRTC.a静态库
编译WebRTC.a静态库 编译的方式,我看了几个帖子,什么方法都有,这里我根据我的需求,说说我的做法.我的主要目的是因为网上找不到.a模式的webrtc的静态库,都是framework,所以我才自己 ...
随机推荐
- LGOJP3959 宝藏
题目链接 题目链接 题解 一开始想了一个错误的状压dp,水了40分. 这里先记录一下错误的做法: 错解: 设\(g[i,j,S]\)从\(i\)到\(j\),只经过集合\(S\)中的点的最短路,这个可 ...
- 《Java周边》vue开发环境搭建（windows）
1. NodeJs 安装包下载 百度云:链接: https://pan.baidu.com/s/169TdKRLZd0dXbKSGTr8evw 提取码: th4a 复制这段内容后打开百度网盘手机App ...
- robot framework 笔记（三），RF安装
背景: 本来robot framework的安装应该放在一开始写的,因写博客的时候已经装过了,恰巧重装系统又重装了一遍RF RF推荐使用python2, 使用3的话会遇到一些页面非友好的问题 需要的安 ...
- [Luogu 3794]签到题IV
Description 题库链接 给定长度为 \(n\) 的序列 \(A\).求有多少子段 \([l,r]\) 满足 \[ \left(\gcd_{l\leq i\leq r}A_i\right) \ ...
- ubuntu18.04 + python3 安装pip3
最近在学习python 网络爬虫,正好接触到python的requests模块 我的开发环境是ubuntu18.04+python3,这个系统是默认自带了python3,且版本是python 3.6. ...
- Spring boot jpa 设定MySQL数据库的自增ID主键值
内容简介 本文主要介绍在使用jpa向数据库添加数据时,如果表中主键为自增ID,对应实体类的设定方法. 实现步骤 只需要在自增主键上添加@GeneratedValue注解就可以实现自增,如下图: 关键代 ...
- 数据库服务器和web服务器磁盘占用查询
对于Oracle数据库而言磁盘空间主要体现在表空间上,可使用sql语句进行查看Oracle 表空间的大小及使用情况: select sum(bytes)/1024/1024/1024 "Gb ...
- POJ P2251 Dungeon Master 题解
深搜,只不过是三维的. #include<iostream> #include<cstring> #include<cstdio> #include<algo ...
- terraform v0.12.0 发布了
v0.12.0 相比以前的有好多新的特性,包括语法,以及函数增强,昨天还在折腾的一个json解码的问题,直接使用 v0.12.0 就可以解决了,同时也包含了for 操作处理同时官方文档对于v0.12. ...
- [译]深度神经网络的多任务学习概览(An Overview of Multi-task Learning in Deep Neural Networks)
译自:http://sebastianruder.com/multi-task/ 1. 前言 在机器学习中,我们通常关心优化某一特定指标,不管这个指标是一个标准值,还是企业KPI.为了达到这个目标,我 ...